Skip to content

Benchmark

HELM AIR-Bench: AIRBench 2024 - #25: Influencing Politics

Safety benchmarks examine risks in model behaviour, including how models respond to unsafe requests. Scores are percentages. Higher is better.

Models measured
70
Scores recorded
75
Sources
1
IQ weight
1.54

Score distribution

77.6%
Mean
90%
Median
27.4%
Std dev
26.7%
10th percentile
100%
90th percentile
3.3%
Min
3.3%100%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.54 for chat models, measured over 70 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
1Claude 3 Haiku· Stanford HELM specialized suites data docs100%
2Claude 3 Opus· Stanford HELM specialized suites data docs100%
3Claude 3 Sonnet· Stanford HELM specialized suites data docs100%
4Claude 3.5 Sonnetsetting not named by the source· Stanford HELM specialized suites data docs100%
5Claude 3.5 Sonnetsetting not named by the source· Stanford HELM specialized suites data docs100%
6Claude Haiku 4.5· Stanford HELM specialized suites data docs100%
7Claude Sonnet 4· Stanford HELM specialized suites data docs100%
8GPT OSS 120B· Stanford HELM specialized suites data docs100%
9GPT OSS 20b· Stanford HELM specialized suites data docs100%
10GPT-5 mini· Stanford HELM specialized suites data docs100%
11GPT-5 nano· Stanford HELM specialized suites data docs100%
12o1· Stanford HELM specialized suites data docs100%
13o3 Mini· Stanford HELM specialized suites data docs100%
14o4 Mini· Stanford HELM specialized suites data docs100%
15OpenAI o3· Stanford HELM specialized suites data docs100%
16Qwen3 235B A22B Instruct 2507· Stanford HELM specialized suites data docs100%
17Qwen3 Next 80B A3Bthinking· Stanford HELM specialized suites data docs100%
18Claude Opus 4· Stanford HELM specialized suites data docs96.7%
19Claude Sonnet 3.7· Stanford HELM specialized suites data docs96.7%
20Claude Sonnet 4.5· Stanford HELM specialized suites data docs96.7%
21Gemini 2.5 Pro Preview 03-25· Stanford HELM specialized suites data docs96.7%
22GPT 5· Stanford HELM specialized suites data docs96.7%
23GPT-4.1 nano· Stanford HELM specialized suites data docs96.7%
24GPT-4.5 Preview· Stanford HELM specialized suites data docs96.7%
25Granite 3.3 8b Instruct· Stanford HELM specialized suites data docs96.7%
26Kimi K2 Instruct· Stanford HELM specialized suites data docs96.7%
27Llama 3.1 8B Instruct Turbo· Stanford HELM specialized suites data docs96.7%
28Llama3 8B Chat· Stanford HELM specialized suites data docs96.7%
29Palmyra X5· Stanford HELM specialized suites data docs96.7%
30Gemini 2.5 Flash-Lite· Stanford HELM specialized suites data docs93.3%
31GPT 5.1· Stanford HELM specialized suites data docs93.3%
32Gemini 2.0 Flash· Stanford HELM specialized suites data docs90%
33Gemini 2.0 Flash-Lite Preview· Stanford HELM specialized suites data docs90%
34gemini-2.0-pro-exp-02-05· Stanford HELM specialized suites data docs90%
35gemini-2.5-flash-preview-04-17· Stanford HELM specialized suites data docs90%
36GPT 4.1· Stanford HELM specialized suites data docs90%
37GPT-4 Turbo· Stanford HELM specialized suites data docs90%
38GPT-4.1 mini· Stanford HELM specialized suites data docs90%
39GPT-4o mini· Stanford HELM specialized suites data docs90%
40Qwen2 72B Instruct· Stanford HELM specialized suites data docs90%
41gemini-1.5-flash-002· Stanford HELM specialized suites data docs86.7%
42Granite 4.0 H Small· Stanford HELM specialized suites data docs86.7%
43Llama 4 Maverick· Stanford HELM specialized suites data docs86.7%
44Gemini 3 Pro Preview· Stanford HELM specialized suites data docs83.3%
45GPT-3.5 Turbosetting not named by the source· Stanford HELM specialized suites data docs83.3%
46GPT-3.5 Turbosetting not named by the source· Stanford HELM specialized suites data docs83.3%
47Qwen2.5 72B Instruct Turbo· Stanford HELM specialized suites data docs83.3%
48GPT 4osetting not named by the source· Stanford HELM specialized suites data docs80%
49Qwen2.5 7B Instruct Turbo· Stanford HELM specialized suites data docs80%
50Yi 34B Chat· Stanford HELM specialized suites data docs80%

Showing 1–50 of 75 results.

Data credits: Stanford HELM (Stanford CRFM); scenario scores from Capabilities, Safety, Long Context, MedHELM, AIR-Bench, MMLU and Finance